Skip to Content
Uncategorized

How Wi-Fi Drains Your Cell Phone

A study finds problems, but also offers software fixes that could help cell phones last significantly longer between charges.
June 24, 2010

Some simple changes to the software running on Wi-Fi access points could significantly extend or even double cell phone battery life. That’s the finding of a study that investigated why using Wi-Fi on a cell phone, and on some other portable devices, sucks up power so quickly. It found that a protocol designed to reduce Wi-Fi power drain often doesn’t work effectively.

Eric Rozner at the University of Texas at Austin and colleagues from the University of Wisconsin-Madison and Microsoft Research India made the discovery, and they also came up with a fix for the problem.

The team began by benchmarking just how much power different models of cell phones needed to use Wi-Fi. “For example, we found that an HTC Tilt’s total power consumption increases by threefold when using Wi-Fi,” says Rozner, who notes that previous studies have shown Wi-Fi use can account for up to 60 percent of the phone’s total energy consumption.

“It is somewhat surprising that Wi-Fi consumes so much energy,” Rozner says. He explains that a protocol called Power Saving Mode exists to prevent Wi-Fi from draining mobile devices’ batteries too quickly. But when the team studied how a variety of access points use this mode, it found that the setup wasted power and unfairly prioritized some devices over others. “We found that current implementations of Power Saving Mode suffer multiple problems,” says Rozner.

Wi-Fi’s hunger for energy is important. “More and more carriers are encouraging their subscribers to reduce 3G usage and instead use Wi-Fi by capping 3G data usage or enforcing certain applications to run exclusively on Wi-Fi,” Rozner explains.

A mobile device using Power Saving Mode flips its wireless radio between fully powered and a sleep setting, for periods lasting between seconds and tens of milliseconds, to conserve energy. For example, after sending a request for a file from the Web, a phone might sleep if it doesn’t receive the file after half a second. While sleeping, the device listens for a beacon message that indicates its data is ready, after which it switches to full power and asks the access point to send it.

That works out fine when an access point is only serving that one device. But in reality, it is likely sending data to other devices, too, such as laptops. When a phone wakes up and requests its data, many access points simply add it to the back of the queue of outgoing packets, even if the phone’s data arrived at the access point long before those in front.

As a result, the phone burns energy while it waits for its data to advance to the head of the line, a situation that saps battery life. Some phones, like the iPhone, won’t wait for more than a few tens of milliseconds and go back to sleep if the data isn’t forthcoming. But this can also waste power, as well as network capacity; when the access point does send it, the phone cannot receive it and must wake up and request it all over again.

Some access points sidestep those problems by bumping Power Saving Mode traffic to the head of the queue, but that can degrade the quality of wireless signal for everyone else. “We witnessed reduced network capacity due to unnecessary retransmissions and unfairness to network traffic,” says Rozner. He and his colleagues developed an alternative way of handling Power Saving Mode traffic that slashes the energy use of mobile devices and maintains a fair playing field for all traffic.

Their system, dubbed NAPman, carefully enforces a first-come, first-served approach to all data, whether it’s from a device using Power Saving Mode or not. It also only wakes a phone to retrieve its data when its data it is at the front of the queue, preventing the phone from waiting around and burning energy. The system also tracks devices that go to sleep after a fixed time, to ensure they aren’t sent data while asleep.

NAPman also uses the ability of Wi-Fi access points to pose as virtual access points to assign different virtual connections to different clients. The result is that devices do not compete for traffic so directly, and the access point can carefully time the sending of its beacons to ensure that devices only wake when necessary.

“Not only could we provide 70 percent energy savings compared to the conventional implementation, but NAPman is fair to background traffic,” says Rozner. In a test that involved streaming a 128 kilobit-per-second radio stream to an HP iPAQ smart phone using a crowded hot spot, NAPman doubled the device’s battery life from 4.7 to 10 hours, although if the backlight was turned up high, the effect was slightly reduced.

“It seems that though systems today may be deploying power saving, they’re doing it wrong,” says Philip Levis, who works on networking at Stanford University, speaking after Rozner presented his work at the MobiSys conference in San Francisco last week. “But I wonder how specific this is to just devices today,” he adds.

Rozner acknowledges that as new devices are released, his fixes may become unnecessary. “But as new schemes are implemented, I think they will need to take into account some of our ideas,” he says. The team hasn’t entered discussions with any wireless router vendors. “But the potential for adaptation is there,” says Rozner.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.